Preview.jpeg

Radiance GeoJSON - Global Night Light Pollution Data 2024

by: Ignacio de Miguel Diaz

Publication date: 2025-11-22
Usage: Attribution 4.0 International
Topics: light pollution, VIIRS, nighttime lights, geospatial data, environmental science, satellite imagery, astronomy, dark sky, bortle scale
Contributor: Earth Observation Group, Colorado School of Mines
Language: English
Rights: Derived dataset licensed under CC BY 4.0. Based on VIIRS Nighttime Lights data by Earth Observation Group, Colorado School of Mines (CC BY 4.0). See https://eogdata.mines.edu/products/vnl/ and https://creativecommons.org/licenses/by/4.0/. Database dump and all transformations are also distributed under CC BY 4.0.
Item Size: 10.3G

Country-level nighttime light pollution dataset derived from VIIRS Nighttime Lights imagery, offering 0.5 km resolution radiance values, MPSAS sky brightness, and Bortle scale classifications for 173 countries.

This collection includes individual GeoJSON files (.json.gz) for each country, plus a fully processed PostgreSQL/PostGIS database dump (custom format, ~4.4 GB compressed, ~27 GB when restored) containing 139,974,408 georeferenced points assembled into a single optimized spatial database. The database includes precomputed Web Mercator geometry (geom_3857) for fast web mapping, spatial GIST indexes on both WGS84 and Web Mercator geometries, and a ready-to-use light_points_tile(z, x, y) function that generates Mapbox Vector Tiles directly from PostGIS—suitable for pg_tileserv or custom tile servers.

Generated using the MIT-licensed radiance-geojson tool by Ignacio de Miguel Díaz, based on VIIRS Nighttime Lights data (CC BY 4.0) from the Earth Observation Group, Colorado School of Mines. The database dump enables direct restoration into a PostGIS instance for advanced geospatial analysis, web tiling, and API backends without additional preprocessing.

Notes

Original VIIRS Nighttime Lights GeoTIFF tiles were provided by the Earth Observation Group, Colorado School of Mines (CC BY 4.0). This item contains processed GeoJSON files (.json.gz) and a PostgreSQL/PostGIS database dump that assembles all spatial features, indexes, and metadata for advanced geospatial analysis and direct database restoration.

The database has been built by batch-loading and extracting every JSON file through PostGIS, offering full spatial querying capabilities for researchers and developers. The optimized dump includes precomputed Web Mercator geometry for fast web mapping, spatial GIST indexes on both WGS84 and Web Mercator coordinate systems, and a vector tile generation function compatible with pg_tileserv and custom mapping applications.

Database structure: main table light_points (139,974,408 rows) with columns for WGS84 location, Web Mercator geometry (geom_3857), radiance values, MPSAS sky brightness, Bortle scale classification, elevation, and country code. Primary key on id, B-tree indexes on radiance and country_code, GIST spatial indexes on both geometry columns. Includes helper function light_points_tile(z, x, y) for generating Mapbox Vector Tiles directly from PostGIS.

Version history: Initial release 2025-11-22 (GeoJSON files). Database dump v1 added 2025-12-01 with optimized geometry and tile function for high-performance vector tile serving.